A novel profluorescent probe for detecting oxidative stress induced by metal and H₂O₂ in living cells

Abstract

A profluorescent probe that has no fluorescent response to H₂O₂, iron or copper ions but can be readily activated in the presence of both H₂O₂ and Fe (or Cu) ion has been developed; the probe is capable of detecting oxidative stress promoted by Fe (or Cu) and H₂O₂ (i.e. the Fenton Reaction conditions) in living cells.

Oxidative stress plays a major role in the pathogenesis of a large number of human diseases, among which, the highly reactive and deleterious oxidizing species produced via the reaction of endogeneous H₂O₂ and redox metals (e.g., iron and copper) (the Fenton reactions (eq. 1)) have been implicated in the pathogenesis of Wilson’s disease, Parkinson’s disease (PD), Alzheimer’s disease (AD), atherosclerosis, hemochromatosis, liver damage, cancer and diabetes, etc.¹ Abnormal accummulation of redox metals (e.g., iron and copper) and overproduction of H₂O₂ in certain tissues in the body have been observed in patients with neurodegenerative diseases.¹ Elevated levels of redox metal ions and H₂O₂ and the Fenton reactions contribute to the oxidative stress and neurodegeneration. We and others have been developing agents capable of performing H₂O₂-triggered “anti-Fenton reaction” via a prochelator activation and a subsequent metal caging strategy.^2,3 The advantage of this novel strategy is that chelation can only be triggered by toxic levels of H₂O₂ thus will not interfere with the healthy metal homeostasis, holding promise in combating these diseases. However, a direct demonstration of the mechanism of this novel strategy in living systems is still lacking.

$${\text{Fe}}^{\text{II}}({\text{Cu}}^{\mathrm{I}})+{\mathrm{H}}_2{\mathrm{O}}_2\to {\text{Fe}}^{\text{III}}({\text{Cu}}^{\text{II}})+{\text{HO}}^{-}+\text{HO}•$$ (1)

To provide direct evidence on the mechanism of this novel anti-Fenton strategy in living systems, we have developed our third generation prochelator, RS-BE, which is profluorescent and does not react with Fe (or Cu) ions. However, it can be converted to an active chelator by H₂O₂, still fluorescence silent, but the fluorescence is activated upon metal chelation on the active chelator. Thus, the “anti-Fenton process” is readily monitored by an ideal “turn-on” fluorescent process in living cells which is described in this communication.

The profluorescent prochelator RS-BE was designed by “masking” the key chelating hydryoxyl group in the active chelator Rh-SBH with a bulky boronic acid pinacol ester group which may be unmasked by H₂O₂ (Scheme 1). Subsequent metal coordination with the active chelator Rh-SBH may induce the conversion of the profluorescent Rh-SBH molecule from the ring-closed spirolactam form (non-fluorescent) to the ring-opened amide form (fluorescent) in the metal complexes (Scheme 1).⁴ This metal-coordination induced profluorescence activation process may be called a “coordination-induced fluorescence activation (CIFA)”.

Scheme 1.

Synthesis and activation of RS-BE

Synthesis of the prochelator RS-BE was accomplished in 3 0 % yield by refluxing rhodamine hydrazine and 2-fomylphenylboronic acid, pinacol ester in ethanol (ESI^†). The active chelator, Rh-SBH, was also synthesized and characterized following a published procedure. ⁴

The UV-vis spectroscopic properties of RS-BE and its interactions with Cu²⁺, Fe²⁺ and H₂O₂ were evaluated first. Due to limited water solubility of RS-BE, a mixed solvent acetonitrile(ACN)/potassium phosphate buffer (KPB) (10 mM, pH 7.5, v/v 1:1) was used. Solution of RS-BE (50 µM) is colorless, exhibiting absorption in UV region (240–300 nm) only (Fig.S1). As shown in Fig.1 and Fig. S1, addition of Cu²⁺ or Fe²⁺ ions changes little the aborption characteristics of RS-BE, suggesting the prochelator does not bind strongly to Cu²⁺ or Fe²⁺ ions under the conditions. However, upon addition of excess H₂O₂ to the systems (60 min incubation with 500 µM H₂O₂), the solutions turned pink and a new peak at 550 nm was observed in both the Cu²⁺- and Fe²⁺-RS-BE systems (Fig.1 and Fig. S1; Fe²⁺ may be oxidized to Fe³⁺ by excess H₂O₂ here), implying metal-chelation occurred via a H₂O₂-triggered prochelator activation (from RS-BE to Rh-SBH) and subsequent metal-chelation mechanism (Scheme 1), as demonstrated for our previously designed prochelators.² Further investigations on the interactions of Fe²⁺ or Fe³⁺ with the active chelator Rh-SBH suggest that Rh-SBH response to Fe³⁺, not Fe²⁺(Fig. S2). A strong response of Rh-SBH to Cu²⁺ but a weaker response to Fe²⁺ were also reported under aerobic conditions in a different buffer (ACN/Tris, 10 mM, pH 7.0).⁴ The appearance of the new peak at 550 nm has been assigned to the metal-binding induced conversion from ring-closed spirolactam form of Rh-SBH (colourless) to the ring-opened amide form (pink) in the metal complexes.⁴ Rh-SBH displays a selective absorption response to Cu²⁺ and Fe³⁺ over other metal ions (Fig. S3), and the spirolactam form is stable over pH 5.0 to 8.2, in agreement with that reported.⁴

Fig. 1.

Absorption spectra of RS-BE and its interactions with Cu²⁺ (I) or Fe²⁺ (II) before and after addition of 500 µM H₂O₂ in ACN/KPB buffer (10 mM, pH 7.5, v/v 1:1): (a) RS-BE (50 µM) only, (b) RS-BE with metal ions (50 µM) and (c) with metal ions and H₂O₂ (500 µM).

The clean conversion of RS-BE to RH-SBH by H₂O₂ was further confirmed via NMR spectroscopy. As shown in Fig. 2, after incubating RS-BE with H₂O₂, ¹H NMR peaks for RS-BE (δ 9.42(s), =CH-) gradually decreased in intensity, while the peaks corresponding to Rh-SBH (δ 9.09(s), =CH-; 10.45(s) – OH−) appeared simultaneously and increased in intensity with time. The peaks (δ7.81(d), 7.51–7.57(m) and 7.30(t), salicylaldehyde) also underwent similar conversions. Meanwhile, the peaks for boric acid and pinacol, the H₂O₂-deprotected products of the boronic acid pinacol ester moiety, appeared at δ6.55(s, B(OH)₃), δ7.99(s, -OH) and δ1.15 (s, 4×(CH₃), Figs. S5, S6). No ¹H NMR change was observed for the xanthene moiety due to its distance from the reaction site. In ~2 h, RS-BE had been cleanly converted to Rh-SBH with no intermediate formed, as indicated by the ¹H-NMR spectra. ¹³C-NMR data also confirmed that RS-BE was converted by H₂O₂ to Rh-SBH which is still in the the ring-closed spirolactam form (Fig. S7).

Fig. 2.

¹H NMR spectra (from δ11.0- 6.0, (CD₃)₂SO) ) of (a) RS-BE (5 mM); (b) and (c), the reaction of RS-BE (5 mM) with H₂O₂ (50 mM) at 293 K after 1 h and 2 h, respectively; and (d) Rh-SBH (5 mM).

We further examined the fluorescence responses of RS-BE to Cu²⁺, Fe²⁺ and H₂O₂. As shown in Fig. 3, RS-BE in ACN/KPB buffer (1:1, pH 7.5) exhibits weak fluoresence at 575 nm (curve (a)). The fluorescence is not affected by the presence of H₂O₂ (Fig. S8). Addition of Cu²⁺ or Fe²⁺ (curves (b) in Fig. 3) to RS-BE in the absence of H₂O₂ did not change its fluorescence profiles either. However, upon the addition of H₂O₂ to the system containing RS-BE/Cu²⁺(or Fe²⁺), as we expected, the fluorescent intensity at 575 nm increased with time (curves (c)–(g) in Fig.3 (I) and (c)–(f) in Fig. 3 (II)). At 2 h, a ~7.5-fold increase in fluorescent intensity was observed in the RS-BE/Cu²⁺/H₂O₂ system while a ~5-fold increase in the RS-BE/Fe²⁺/H₂O₂ system (Fe²⁺ may be oxidized to Fe³⁺ by excess H₂O₂ here). In addition, the emission spectrum of RS-BE/Cu²⁺(or Fe²⁺) underwent a red shift from 575 nm to 580 nm after the addition of H₂O₂. This fluorescence response matches those observed for the reactions of Rh-SBH with metal ions.⁴ Therefore, we can conclude that in the presence of Cu²⁺(or Fe²⁺), the fluorescence enhancement of RS-BE upon additon of H₂O₂ must be occurred via a H₂O₂-triggered prochelator activation (from RS-BE to Rh-SBH) and subsequent metal-chelation mechanism (Scheme 1), corroborating the conclusions from the UV-vis studies (Fig. 1). These interesting fluorescent responses offer us the opportunity to study the “anti-Fenton” mechanism in living cells by using RS-BE as a novel “turn-on” fluorescent probe.

Fig. 3.

H₂O₂-triggered fluorescence responses (E_x, 510 nm; E_m, 580) of RS-BE (50 µM) to Cu²⁺ (I) or Fe²⁺ (II) (50 µM) in ACN/KPB buffer (10 mM, pH 7.5, v/v 1:1). (Ia) and (IIa) RS-BE only; (Ib) and (IIb) RS-BE with Cu²⁺ and Fe²⁺, respectively; from (Ic) to (Ig), RS-BE/Cu²⁺ incubated with 500 µM H₂O₂ for 10, 30, 60, 90 and 120 min, respectively; from (IIa) to (IIf), RS-BE/Fe²⁺ incubated with 500 µM H₂O₂ for 30, 60, 90 and 120 min, respectively.

RS-BE was then tested in live cells (SH-SY5Y neuroblastoma cells, a human neuronal cell line) its fluorescent responses to Cu²⁺, Fe³⁺ and H₂O₂ via a laser scanning confocal microscope (Zeiss LSM 710). Iron- and copper-8-hydroxylquinoline complexes (Fe(8-HQ) and Cu(8-HQ)) were used to enhance membrance permeability of the metal ions.^5,6 Human SH-SY5Y cells loaded with 10 µM RS-BE (incubated for 30 min) did not show intracellular fluoresence (Fig. 4(b)). Then, the cells were further implemented with 10 µM Fe(8-HQ). No fluorescence was observed after 30 min incubation (Fig. 4(c)), suggesting that the normal cellular components or the added Fe(8-HQ) does not trigger fluorescent signal of the profluorescent probe RS-BE, as we desired. At last, we treated the Fe(8-HQ) loaded cells with 100 µM H₂O₂, followed by immediate recording fluorescence images of the cells every 5 mins for 30 min. Excitingly, as shown in Fig. S9 and Fig. 4(d) and (e), intracellular fluorescent signal which matches the profile of Rh-SBH-metal comlexes emerged after 5 min treatment with H₂O₂ and increased in intensity with time. A ~10-fold increase in intensity was observed after 25 min (Fig.4(f)). These data suggest that a H₂O₂-triggered prochelator activation (from RS-BE to Rh-SBH) followed by Fe-chelation occurred in the cells.

Fig. 4.

Confocal fluorescence images of live human SH-SY5Y cells with the treatment of RS-BE/Fe/ H₂O₂ (scale bar 10 µm). (a) DIC; (b) the cells incubated with 10 µM RS-BE for 30 min; (c) the cells were then incubated with 10 µM Fe(8-HQ) for 30 min; (d) and (e) the cells were further treated with 100 µM H₂O₂ for 10 and 25 min, respectively, (f) Integrated emission (547–703 nm) intensity of (a), (b), (c), (d) and (e) images.

Similar experiements were performed to test the response of RS-BE with Cu²⁺ and H₂O₂ in human SH-SY5Y cells. As expected, supplement of 10 µM RS-BE and subsequent 10 µM Cu(8-HQ) or CuCl₂ did not result in an intracellular fluorescent response (Fig.5(b) and (c), Fig. S10). Further addition of 100 µM H₂O₂ to the cells loaded with RS-BE and Cu(8-HQ) displayed a specific intracellular fluorescence enhancement (Fig. 5(d) and (e)) but with weaker intensity compared to those with Fe³⁺. In 60 min, only ~3.8-fold increase in intensity was observed. Little intracellular fluorescence enhancement was observed 30 min after the addition of 100 µM H₂O₂ to the cells preloaded with RS-BE and CuCl₂ (Fig. S10). However, cells treated with higher levels of H₂O₂ (500 µM) and CuCl₂ (50 µM) did exhibit the specific intracellular fluorescence enhancement, though still weak (Fig. S11). The weaker fluorescent response in the cells loaded with Cu²⁺ may be due to poor availability of free copper ions in cells.⁷

Fig. 5.

Confocal fluorescence images of live human SH-SY5Y cells with the treatment of RS-BE/Cu/ H₂O₂ (scale bar 10 µm). (a) DIC; (b) cells incubated with 10 µM RS-BE for 30 min; (c) the cells were then incubated with 10 µM Cu(8-HQ) for 30 min; (d) and (e) the cells were further treated with 100 µM H₂O₂ for 30 and 60 min, respectively; (f) Integrated emission (547–703 nm) intensity of (a), (b), (c), (d) and (e) images.

Taken together, we have developed a novel prochelator-type profluorescent sensor RS-BE that does not have fluorescent response to H₂O₂, iron or copper ions but the fluorescence can be readily “turned-on” in the presence of both H₂O₂ and Fe (or Cu) ions. The sensor works via a H₂O₂- triggered prochelator activation (from RS-BE to Rh-SBH) followed by metal-coordination-induced fluorescence activation (CIFA) mechanism, thus providing an ideal probe to monitor the “anti-Fenton” processes by a “turn-on” fluorescence process. This sensor has been demonstrated the ability to monitor the in situ presence of H₂O₂ and Fe (or Cu) ions (i.e. the “Fenton Reaction” conditions) in live human SH-SY5Y cells, therefore, is capable of detecting oxidative stress promoted by H₂O₂ and Fe (or Cu) in cellular system. However, upon being “turned on”, this sensor cannot signal the subsequent drop in H₂O₂ levels as the reaction of RS-BE with H₂O₂ to generate Rh-SBH is irreversible.